Nightly 0.1.0.dev20250711

📗 Getting started

Install via pip

pip install tt_forge_fe==0.1.0.dev20250711 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/
pip install tt_tvm==0.1.0.dev20250711 --pre --extra-index-url https://pypi.eng.aws.tenstorrent.com/

Docker container

ghcr.io/tenstorrent/tt-forge-fe-slim:0.1.0.dev20250711

More detailed instructions can be found in the Getting Started docker section

Test status: ✅

Workflow run: https://github.com/tenstorrent/tt-forge-fe/actions/runs/16225230338

🎯 Other Changes

• Implemented reshape op in cpp. (#2481) by @acolicTT
• Remove xfail marker from stereo pytorch and address monodepth pcc issues (#2479) by @chandrasekaranpradeep
• Remove 'deprecated_verification' from the test plan (#2434) by @vobojevicTT
• Uplift third_party/tt-mlir to origin/main (42332a2b361f458595be7b91769a17b9a92b2fc7) 2025-07-10 (#2492) by @vmilosevic
• Simplified OpType attributes usage. (#2473) by @acolicTT
• Remove model analysis docs generation pipeline (#2487) by @chandrasekaranpradeep
• [CI] Increase num test groups for BH nightly (#2497) by @nsumrakTT
• Fix onnx llama, phi and qwen models (#2493) by @chandrasekaranpradeep
• [OP] Divide migrated to cpp (#2482) by @vkovinicTT
• [cleanup] remove torch 2.0 integration (#2501) by @pilkicTT
• Run perf benchmark in tt-forge (#2452) by @nsumrakTT
• [deps] bump version of (#2499) by @pilkicTT
• [OP] Sine / Cosine migrated to cpp (#2514) by @mstojkovicTT

Files:

tt_tvm-0.1.0.dev20250711-cp310-cp310-linux_x86_64.whl

tt_forge_fe-0.1.0.dev20250711-cp310-cp310-linux_x86_64.whl

Model Compatibility

frontend	model_name	Wormhole
tt-forge-fe	pt_vilt_dandelin_vilt_b32_finetuned_vqa_qa_hf	✅
tt-forge-fe	pt_albert_base_v1_token_cls_hf	✅
tt-forge-fe	pt_albert_textattack_albert_base_v2_imdb_seq_cls_hf	✅
tt-forge-fe	pt_bert_emrecan_bert_base_turkish_cased_mean_nli_stsb_tr_sentence_embed_gen_hf	✅
tt-forge-fe	pt_autoencoder_linear_img_enc_github	✅
tt-forge-fe	pt_efficientnet_efficientnet_b0_img_cls_timm	✅
tt-forge-fe	pt_ghostnet_ghostnet_100_img_cls_timm	✅
tt-forge-fe	pt_hrnet_hrnet_w18_small_v1_pose_estimation_osmr	✅
tt-forge-fe	pt_mobilenetv1_basic_img_cls_torchvision	✅
tt-forge-fe	pt_mobilenetv2_basic_img_cls_torchhub	✅
tt-forge-fe	pt_resnet_50_img_cls_hf	✅
tt-forge-fe	pt_resnext_resnext101_32x8d_img_cls_torchhub	✅
tt-forge-fe	pt_segformer_nvidia_mit_b0_img_cls_hf	✅
tt-forge-fe	pt_vit_google_vit_base_patch16_224_img_cls_hf	✅
tt-forge-fe	pt_vovnet_vovnet27s_img_cls_osmr	✅
tt-forge-fe	onnx_efficientnet_efficientnet_b0_img_cls_timm	✅
tt-forge-fe	onnx_mobilenetv2_mobilenetv2_100_img_cls_timm	✅
tt-forge-fe	pt_albert_base_v1_mlm_hf	✅
tt-forge-fe	pt_albert_base_v2_mlm_hf	✅
tt-forge-fe	pt_albert_base_v2_token_cls_hf	✅
tt-forge-fe	pt_bert_bert_base_uncased_mlm_hf	✅
tt-forge-fe	pt_bert_phiyodr_bert_large_finetuned_squad2_qa_hf	✅
tt-forge-fe	pt_distilbert_distilbert_base_cased_mlm_hf	✅
tt-forge-fe	pt_perceiverio_deepmind_language_perceiver_mlm_hf	✅
tt-forge-fe	pt_ghostnet_ghostnet_100_in1k_img_cls_timm	✅
tt-forge-fe	pt_mobilenetv3_mobilenet_v3_large_img_cls_torchhub	✅
tt-forge-fe	pt_resnext_resnext50_32x4d_img_cls_torchhub	✅
tt-forge-fe	pt_wideresnet_wide_resnet50_2_img_cls_torchvision	✅
tt-forge-fe	pt_xception_xception71_tf_in1k_img_cls_timm	✅
tt-forge-fe	onnx_resnet_50_img_cls_hf	✅
tt-forge-fe	onnx_segformer_nvidia_mit_b0_img_cls_hf	✅
tt-forge-fe	tf_resnet_resnet50_img_cls_keras	✅